Economic and Fiscal Policy Determinants of

Public Deficits: The Philippine Case

By

Benjamin E. Diokno1
I. INTRODUCTION
The Philippine national government experienced large and unsustainable budget deficits
in the 1980s. After a brief period of near balanced budget in the mid-1990s, large budget
deficits have reemerged in recent years. But unlike the heavy fiscal imbalances in the
early 1980s which were caused by large investment in public infrastructure and low tax
effort, the return of large fiscal deficits in recent years was accompanied by falling tax
effort and underspending for education, health and public infrastructure. With deficits
rising and investment in human capital and public infrastructure deteriorating, an
appropriate question is: what has caused the poor fiscal performance of the Philippines in
recent years? Is it the result of unfortunate events, macroeconomic shocks or misdirected
fiscal policy?

Chart 1: Fiscal Performance: 1981-2005

As percent of GDP

Large deficits have reemerged

25.0
20.0
15.0

Rev

10.0

Exp

5.0

Deficit
2004

2002

2000

1998

1996

1994

1992

1990

1988

-10.0

1986

-5.0

1984

0.0

This paper estimates how macroeconomic and fiscal policy variables affect the fiscal
position of the government. There are at least three possible ways of measuring the fiscal
1

I wish to acknowledge the Philippine Center for Economic Development for its financial support and
Justine Diokno-Sicat for her research assistance. The views expressed here are those of the author and are
not necessarily those of the Philippine Center for Economic Development and the School of Economics of
the University of the Philippines.
1

health of the Philippines: the national government fiscal balance (NGFB or NGAB for
national government account balance), the consolidated public sector financial position
(CPSFP or CPSD for consolidated public sector deficit), or the public sector borrowing
requirement (PSBR). The NGFB or NGAB which measures the fiscal performance of the
national government alone is the one generally understood by policymakers, the media
practitioners and the general public. Among the three measures, NGAB is no doubt the
narrowest and the least accurate in describing the true fiscal position of the government.
The CPSD, on the other hand, is the combined deficits of the national government, the
monitored government-owned and controlled corporations, government financial
institutions, local governments, and other public sector entities. CPSD is a better measure
of the public sectors true state of finances than NGFB. From the economic standpoint,
PSBR is perhaps the most relevant measure of fiscal imbalance. It is the deficit of the
national government and the 14 monitored corporations less the budgetary assistance to
the monitored corporations in the form of equity contributions and net lending. It
measures the amount the government has to borrow domestically or externally to finance
the combined deficits of the national government and the monitored state corporations.
Chart 2: Deficit Measures: NGAB, PSBR, CPSFP
In percent of GDP

Fiscal deficits have worsened in recent years

2.0

19
85
19
87
19
89
19
91
19
93
19
95
19
97
19
99
20
01
20
03
20
05

0.0
-2.0
-4.0
-6.0
-8.0
NGFB

CPSD

PSBR

In the paper, two indicators of fiscal health will be used, namely: the national government
account balance (NGAB) and the consolidated public sector fiscal balance (CPSF). The
results of the econometric work will show that fiscal policy variables have more
influence on Philippine fiscal balances compared to macroeconomic variables.
The paper is organized as follows: Section 2 presents a selected review of literature of
theory of fiscal deficits, Section 3 discusses the methodology and data sources and
Section 4 analyzes the results. The final section discusses some implications for policy.

II. THEORETICAL UNDERPINNINGS

What explains persistent budget deficits? What are the theoretical implications of
persistent budget deficits? Theories of budget deficits run in two general directions.
Some theories look at the effect of fiscal deficits on economic variables. Others look at
the reverse direction, that is, what macroeconomic and fiscal variables (including budget
rules and institutions) affect and determine fiscal deficits. This section gives a brief
review of the theories of budget deficits with focus on both macroeconomic and fiscal
policy variables.
A. How Persistent Deficits Affect the Economy

How do persistent budget deficits and large government debt affect the economy?
Macroeconomic theory has divergent hypotheses regarding the implications of
government deficits and debt on the economy. One strand of the literature contends that
government debt reduces national saving which, in turn, crowds out capital
accumulation.2 Thus, government debt hinders economic growth. Another strand of the
literature implies the opposite: public debt does not influence national saving or capital
accumulation. This view is based on the Ricardian equivalence theorem that asserts that
it is only the quantity of government purchases, not whether such purchases are financed
through between taxation or borrowing, which affects the economy. This implies that
economic agents are indifferent between government borrowing now or to a tax increase
in the future.3 It has been shown empirically that this is not the case in the real world. In
addition, when the permanent income hypothesis and the effect on consumption are
considered, the Ricardian equivalence may not hold.4
Barros tax-smoothing theory states that what determines the deficit is the desire of
government to minimize distortions associated with raising taxes. The model implies that
deficits and surpluses arise when the ratio of government purchases to output is expected
to change. War and recession are times when the expected future ratio of government
purchases to output is less than the current ratio. Consistent with the tax smoothing
model, it has been observed that government usually run deficits during these times.5
This implies that when national income is low, or government purchases are large,
governments run deficits.
Roubini and Sachs [1988] find only partial evidence to support tax-smoothing, wherein
tax rates are set over time to minimize the excess burden of taxation. They found a
2

Mankiw [2000]
Ibid
4
Romer [2001]
5
Ibid.
3

tendency for larger deficits in countries characterized by a short average tenure of

government, the presence of many political parties in a ruling coalition and higher tax
collection cost.
B. What Macroeconomic Variables Affect Fiscal Balance?

Inflation may affect budget deficits through various ways. The first way is through real
tax revenues -- inflationary conditions reduce the real tax revenues collected by
government, thus, pushing toward budget deficits. The second way is via the effect on
nominal interest rates. Inflation increases the nominal interest rates and consequently
debt servicing, thus increasing the budget deficit. With these two factors in mind, it may
be expected that inflation negatively affects fiscal balances.6
However, inflation may positively affect fiscal stance by raising revenues via income tax
bracket creep. The US experience in the late 1970s was high federal tax receipts as a
percentage of GDP in the face of high inflation rates (of approximately 10%). The
explanation given by Saez [1999] and Auerbach [2000] was that the US income tax
system at the time was not indexed for inflation (i.e. fixed in nominal terms), resulting in
taxpayers near the top-end of a bracket to creep to the next bracket even if real income
remained the same. Furthermore, if the tax system is designed to be elastic to changes in
economic activity, it may be possible to have increased revenues with a boom and thus a
positive influence on fiscal balance.
Easterly and Schmidt-Hebbel [1994] estimated the relationship between inflation and
fiscal deficits. Across countries, the decision to print money to finance deficits (i.e.
seignorage) would depend on the extent to which other means of financing are available.
In their cross section estimation, they found no simple relationship between fiscal deficits
leading to inflation. For case studies using time series data, revenue-maximizing
inflation rates seem to rise with actual average inflation. In addition, money demand and
inflation are nonlinearly related. It was found that money demand has decreasing semielasticity with respect to inflation. This implies that as inflation rises money demand
becomes less semi-elastic. They concluded that seignorage is unimportant as a steadystate phenomenon, but it can be important as a temporary source of revenue in times of
crisis. Furthermore, large surges of money creation are not closely linked to accelerated
inflation. Though Easterly and Schmidt-Hebbel [1994] looked at how budget deficits
affect inflation via seignorage, the opposite direction of this study, it is evident that the
relationship of inflation and fiscal stance is not a simple one. The effect of inflation may
be through various routes, thus making the actual relationship dependent on empirical
evidence.
The level of development of the financial market is also believed to be related to fiscal
performance. A more developed financial market would have more readily available
forms of money to buy goods and services without incurring costs. The World Bank
6

Dornbush et al. [2003]

4

suggests that a more developed financial sector has increased flexibility in adjusting to
macroeconomic shocks to prevent banking or financial crises. A measure of financial
depth used by the World Bank is the ratio of liquid liabilities (i.e. broad money or M3) to
GDP.7
Another aspect of a financially deep economy is the link between banking openness and
economic growth. Bayraktar and Wang [2006] found empirical evidence that banking
sector openness may directly affect growth by improving the access to financial services
and indirectly by improving the efficiency of financial intermediaries, both of which
reduce the cost of financing and in turn, stimulate capital accumulation. Increased
investments lead to economic growth and an improved fiscal performance, implying a
positive relationship.
The literature on financial openness has also hinted at a positive relationship between
financial depth and fiscal balance. Financial repression, as indicated by a less liquid
banking sector, is practiced by government either to finance its budget deficits or to direct
its access of cheap credit to select industries, or both.8 Restrictive financial policy can be
implemented in various ways: (1) imposing high nominal interest rate ceilings; (2) money
creation (i.e. seignorage); and (3) imposing high reserve requirements.9 Denizer, Desai
and Gueorguiev [1998] found evidence that the post-Communist governments in their
study inhibit the development of financial institutions to ensure adequate flows of
external capital to enterprise sectors rather than to finance deficits.
Other empirical evidence, however, has shown a negative relationship between fiscal
deficit and financial market development. Woo [2001] looked at the effect of financial
depth on consolidated public sector deficit in developing countries. He found that an
increase in financial depth is negatively associated with fiscal stance. He explained that a
more liquid banking system can more easily finance fiscal deficits by issuing bonds
without having to resort to inflationary finance. Aizenman and Noy [2003] found similar
evidence that a budget surplus has a negative impact on financial openness for
developing countries. That is, a bigger budget deficit will increase de facto financial
openness. This was explained by evidence that developing economies engage in procyclical, rather than counter-cyclical, policy. In developing economies, financial crises
tend to lead to recessions that in turn result in lower budget deficits because government
reduces its spending. In addition, if the tax system is relatively inelastic to economic
activity, an economic recesiion would lead to relatively higher tax revenues. However, in
the same study, evidence of a positive relationship of fiscal balance and financial
openness was found for OECD countries.
Turning to the open economy, most of the literature and studies about fiscal deficits and
exchange rates have used fiscal stance as the independent variable. Easterly and
Schimdt-Hebbel [1994] found robust relationships between the fiscal deficit, the trade
deficit, and the real exchange rate. The fiscal deficit and the real exchange rate have a
7

2005 World Development Indicators.

Mishkin [2004]
9
Remolona [1985]
8

two-step relationship: the fiscal deficit and other determinants of investment and saving
behavior determine the external deficit, which then determines the real exchange rate
consistent with clearing of the domestic goods market.10 Clarida and Prendergast [1999]
estimated the dynamic relationship between fiscal policy and the real exchange rate in G3
countries since the advent of floating exchange rates. They found that in response to a
fiscal expansion, there is, initially, an appreciation of the domestic currency. However,
over time, the exchange rate overshoots and depreciates relative to the initial rate
prevailing prior to the fiscal expansion shock.
The 1997 Asian financial crisis which was triggered by the collapse of the Thai baht
brought about, through contagion effect, the sharp depreciation of all Asian currencies,
including the overvalued Philippine peso, and an economic slowdown in the region. The
combined effect of the depreciation of the peso, capital flight and decrease in economic
activity contributed to the deterioration of the Philippine fiscal stance starting in 1998.
The negative impact on the Philippine public finances may be attributable to three major
factors: first, higher debt servicing; second, lower revenues because of slower economy
and lower net taxable income of Philippine banks and other private firms; third,
slowdown in economic activity which resulted in lower direct and indirect taxes. On debt
servicing, over the past 25 years, foreign debt has averaged about 50 percent of total
outstanding Philippine government debt.
C. What Fiscal Policy Variables Affect Budget Deficits?

Government has two main policy instruments that are used to direct the economy to a
path of growth and development. First is expenditure policy that is embodied in the
national budget which reflects the spending priorities of government. The second policy
lever is its revenue policy or tax program. In addition, the mode of financing the deficit
has an impact on future fiscal stance.
Expenditure policy
Economic growth theory emphasizes the importance of capital accumulation in the
attainment of economic growth -- the higher the stock of capital the higher the level of
economic output in the long-run. Governments invest in physical infrastructure in order
to increase the productive capacity of an economy. Government spending on public
infrastructure reduces transactions cost for businesses and signals the commitment of
government to ensure profitability for prospective investors. In a study by the World
Bank, Philippine investment in physical infrastructure for the year 2005 was less than 2%
of GDP. This amount is considerably lower than the World Bank prescribed 5% of GDP
to lead to a sustainable economic growth.11

10
11

Rodriguez [1989] as cited in Eaterly & Schmidt-Hebbel [1994].

World Bank [2005].
6

Another policy direction that is believed to have an effect on national government

financial health is fiscal decentralization. The theory of local public good12 argues that
efficiency is enhanced through a process by which constituents reveal their true
preferences for local public goods by voting-with-their-feet, i.e. citizens move to the
locality that offers their most preferred taxing-expenditure mix. Fiscal decentralization
would allow the national government to focus on broader issues such as
interjurisdictional externalities and income redistribution.
The 1991 Local Government Code of the Philippines was enacted in with the aim of
creating self-reliant local government units (LGU). In theory, local authorities are
believed to be more attuned to their constituents and make decisions based on the
preferences of their local constituencies. Moreover, increased spending and revenueraising responsibilities for LGUs enhances accountability. In general, there is a mismatch
between revenue-raising and spending responsibilities, owing to variations in the tax base
and the unequal distribution of income across LGU; this provides the rationale for
intergovernmental fiscal transfers (IGFTs). In the Philippines, the IGFTs -- called internal
revenue allotment (IRA) -- is largely an unconditional block grant, except for 20% which
is required to be allocated to development purposes. The total IRA is 40% of all internal
revenue, based on actual collections in the third preceding fiscal year. The expected
relationship of IRA and fiscal balance is positive since decentralization was designed
create self-reliant local governments.
Revenue policy
Tax revenue is a crucial factor in reducing the probability of persistent budget deficits.13
Auerbach [2003] found that the United States economic downturn, beginning in March
2001, was because of the decrease in federal government revenues rather than increased
spending. About 28% of the loss in projected revenues for 2003 is attributed to new
legislation, i.e. the Bush tax cut; the remaining loss in revenues is attributed to economic
and technical adjustments.
One of the objectives of the study is to quantify the effect of tax reforms on fiscal
balances, via tax effort.14 In the case of the Philippines, the period under study includes
two major tax reforms, namely: (1) 1986 Tax Reform Program and (2) 1997 National
Internal Revenue Code, or Comprehensive Tax Reform Program (CTRP). Diokno [2005]
argues that while the 1986 tax reform program contributed significantly to fiscal
improvements in the late 1980s and early 1990s, the 1997 CTRP was a major contributor
for the progressive decline in tax effort. The peaks and troughs of tax and revenue efforts
in the Philippines are shown in Chart 3 below. The 1986 tax reform program resulted in
higher tax effort which peaked in 1997. Attempts were made to improve upon this tax
performance by tinkering with the tax system in 1997. What came out of the legislature
was a watered-down version of the original proposal. Congress failed to include the
12

For a full discussion on this concept, see the seminal work by Tiebout [1956].
Tax effort is defined as total tax revenue as a percent of GDP.
14
Initially, regressions were run using dummy variables for each major tax reform program. However,
there was a problem of multicollinearity. The tax reform dummy variables are used as instrumental
variables for tax effort instead.
13

crucial rationalization of fiscal incentives and broadening of the value-added tax base.
The reason for this unwanted outcome was the delay in the approval of the 1997 CTRP
tax proposals and the subsequent posturing of politicians who were then aspiring to run in
the 1998 national and local elections.15
Chart 3: Tax Effort and Revenue Effort
As percent of GDP

Tax system: inadequate, unresponsive

25.0
20.0
15.0

Total Rev

10.0

Taxes

5.0

19
84
19
87
19
90
19
93
19
96
19
99
20
02
20
05

0.0

Financing the deficit

If revenues are inadequate to fund planned expenditures, the government has three
options to finance the budget gap: borrow, print money, or increase taxes. In the past,
the Philippine government has resorted to external and domestic borrowing to finance its
deficits. It has amassed huge public debt not only to finance previous years budget
deficits but also to pay for losses incurred by other public sector institutions such as
poorly performing government owned or controlled corporations, public financial
institutions and the Central Bank (CB) but which were later assumed by the national
government.
Government borrowing can crowd out investments in two ways. First, if borrowing is
largely domestic, this may lead to lower investment because of less loanable funds
available for private investors, and thus, to lower output and consumption in the longrun.16 Second, if debt was incurred to settle other debt rather than to finance government
projects in human and physical infrastructure, then crucial public spending is being
forgone. The financing of debt negatively affects important public investment
spending.17 A study by the Asian Development Bank [2005] looked at the implications
15

For a more detailed comparison of the two tax reform programs, see Diokno [2005].
Stiglitz [2000].
17
Diokno [1995].
16

of the current Philippine fiscal policy on government debt. It concluded that the
government debt situation is not sustainable given the current policy regime.
Furthermore, it found evidence of a weak debt Ponzi game.18 This implies that the
Philippine government is simply borrowing to pay off its current debts. Current
government debt is vulnerable to adverse shocks and simple budgetary deficit control
policy is inadequate. These considerations imply a negative relationship between debt
servicing and fiscal balance.
Woo [2001], using panel data, found that debt servicing costs are insignificant
determinants of fiscal deficits.19 Panel data, however, is subject to huge variations and
inconsistently reported data for various countries. It is hypothesized that time-series data
for the Philippines may give different results.
III. METHODOLOGY AND DATA

The objective of this section is to test the relationship of fiscal deficits and the variables
discussed earlier. The estimation method is two-stage least squares method (2SLS).20
Data will be culled from official Philippine government publications.21
A. The Regression Model
FBt = 1 ECONGRt + 2 INFLATt + 3 M 3GDPRATt + 4 REERt
+ 5 INTGDP + 6 CAGDP 2 t + X t + t
where t denotes the year.22
The dependent variable, fiscal balance (FB), represents the two fiscal indicators that will
be presented as a percentage of gross domestic product (GDP). First is the national
government account balance (NGAB) which is defined as total revenues less total
disbursements for the Philippine national government for any given fiscal year. The
second measure of fiscal health is the consolidated public sector fiscal position (CPSFP).
The CPSFP gives an overall view of the public sector. It is the combined surplus (deficit)
of the national government, the Central Bank restructuring accounts, the major nonfinancial government corporations, the government financial institutions, local
government units, the social security institutions (Social Security System, Government
Service Insurance Systems), the Oil Price Stabilization Fund and the Bangko Sentral ng
18

A government is playing a Ponzi game when it keeps on paying old debts with new ones; see Duo Qin,
et. al. [2005].
19
Woo [2001].
20
Initial regressions were run using the ordinary least squares method, however, there was evident
multicollinearity with several variables.
21
See Appendix A for a detailed description and sources of the data.
22
The benchmark regression equation is like that of Woo [2003].
9

Pilipinas (BSP). Separate regressions will be run with each of these as the dependent
variable. In addition, regressions will be run with and without the variable Capital
Outlays as percent of GDP, with a two-year lag.23

B. Macroeconomic Variables

The model includes explanatory variables that are traditionally macroeconomic in nature.
The first variable is the growth rate of real GDP (ECONGR). It is expected that increases
in real GDP growth rates have a positive effect on fiscal balance.
The second independent variable is the rate of growth of the consumer price index (CPI)
or the inflation rate (INFLAT) It may be either positively or negatively related to fiscal
stance so this is an empirical question in the case of the Philippines.24
The variable M3GDPRAT is a proxy for the level of development of the financial
market. It is the ratio of liquid liabilities of the financial system (M3) to GDP. Liquid
liabilities are defined as the sum of currency and deposits in the central bank (M0); plus
transferable deposits and electronic currency (M1); plus time and savings deposits,
foreign currency transferable deposits, certificates of deposit, and securities repurchase
agreements (M2); plus travelers checks, foreign currency time deposits, commercial
paper, and shares of mutual funds or market funds held by residents.
Empirically, Woo [2003] found a significantly negative relationship between liquidity
and fiscal balance. Aizenman and Noy [2003] also found a negative relationship between
financial openness and budget stance for developing countries. In the latter study, a
budget deficit induces financial openness. For OECD countries, they found a positive
relationship, i.e. a budget deficit reduces financial openness. This study would like to
establish empirically the relationship between domestic liquidity and fiscal balance.
The variable REER represents the nominal effective exchange rate of the peso adjusted
for inflation rate differentials with the countries whose currencies comprise the nominal
effective exchange rate basket (NEER1).25 Considering that foreign debt account for
about half of total outstanding government debt, it is anticipated that there is a negative
relationship between real exchange rates and fiscal balance. That is, a depreciation of the
real exchange rate is associated with a worsening fiscal balance.

23

See discussion below.

Dornbusch et al. [2003].
25
Formally, REER = NEER1 x (domestic price index / weighted price index of countries whose currencies
are in the NEER1 basket). The Bangko Sentral ng Pilipinas (BSP) defines NEER1 as the weighted
average exchange rate of the peso vis--vis a basket of foreign currencies (i.e. US Dollar, Japanese Yen,
European Monetary Unit (Euro), British Pound) unadjusted for the effects of inflation. The reason for
using the real exchange rate instead of the nominal exchange rate is to address the problem of
multicollinearity detected from earlier regressions using nominal exchange rates.
24

10

C. Fiscal Policy Variables

Debt servicing is an important variable in examining an economys fiscal position. The
variable INTGDP, is defined as interest payments as percent of GDP for year t. The
expected sign is negative, implying as debt servicing increases (decreases), fiscal balance
worsens (improves).
Traditional growth theories have policy implications for capital stock accumulation. This
study uses capital outlays as a percent of GDP, lagged by two years (CAGDP2) to
quantify its effects on the fiscal balance.26 Capital outlays are lagged by two years to
allow for full completion of infrastructure projects. Regressions are run for the two
specifications with and without CAGDP2. It is expected to be positively associated with
fiscal balance.
The variable Xt represents other explanatory fiscal policy variables.
Another important policy reform that took place during the period under study was the
1991 Local Government Code that seriously promoted fiscal decentralization. Local
governments were assigned more expenditure responsibilities and broader taxing powers.
The intergovernmental transfer system was radically changed: the level of support was
significantly increased, the allocation structure became more predictable and transparent,
and the release procedure was made automatic. The allocation to each level of local
governments provinces, cities, municipalities and barangays is determined through a
formula based on population, land area and equal sharing. The aggregate national
government fiscal transfers to local governments are captured by the variable, IRAEXP,
defined as total internal revenue allotment (IRA) as percent of total national government
spending. The contention is that, with the goal of creating self-reliant local government
units, decentralization efforts have enhanced the efficiency of the delivery of public
goods and services. The expected relationship is positive since LGUs have been given
broader revenue-raising capabilities, somewhat relieving the national government of
financial burden except for IRA.
TAXEFFORT is defined as tax revenues as a percent of GDP. It is expected to have
positive relationship with fiscal balance: the higher (lower) the tax effort, the higher
(lower) the fiscal balance. What is interesting is the effect of tax reforms on tax effort.
The dummy variable TAXREF86 indicates the years in which the 1986 tax reform
program was in force; it is formulated as 1 for all 1986 and onward yeasrs and 0
otherwise since it was not entirely repealed with the CTRP. It is expected that
TAXREF86 has a positive effect on fiscal position since it is broader, simpler to
implement, and more buoyant than the tax system prior to 1986. The dummy variable
TAXREF97 is for the years when the CTRP is in effect. However, this variable also

26

Though capital outlay is also a fiscal policy variable, the manner in which the regressions were run merit
its inclusion in the general economic variables.
11

represents laws that were passed during the Ramos administration that resulted in the
narrower tax base and reduced the tax buoyancy.27
The regressions were run for both specifications (i.e. with and without CAGDP2) using
the tax reform dummy variables in two ways. First, both tax reform dummy variables
was used as instrumental variables for TAXEFFORT. Second, only the tax reform
dummy variable for the CTRP, TAXREF97, was used as an instrumental variable. A
regression for TAXEFFORT was run using both tax reform dummy variables as
independent variables. The expectation is that TAXREF86 would be positively related to
TAXEFFORT while TAXREF97 would be negatively related.28

IV. RESULTS AND ANALYSIS

The results of the empirical estimation for both fiscal balance indicators NGAB and
CPSD -- are presented in this section.
The statistically significant determinants of NGAB are the following: inflation,
domestic liquidity, capital outlays, and tax effort. On the other hand the following
variables were found to be statistically insignificant: economic growth, REER,
interest payment as percent of GDP, and intergovernmental grant (IRA) as percent
of total government expenditures.
Inflation (INFLAT) is shown to have a positive relationship with the national government
account balance (NGAB) because revenues adjust with inflation (that is, higher inflation
leads to higher nominal tax base and consequently higher revenues) while expenditures
are, in general, restricted to what Congress has authorized the President based on the
General Appropriations Act. 29
Domestic liquidity (M3GDPRAT) is shown to have a negative relationship with the
national government account balance. This is in line with previous findings (i.e. Woo
[2003]) implying an increase in domestic liquidity is associated with an increase in the
budget deficit. The availability of non-inflationary finance through more developed
financial system may lead to higher (lower) budget deficit (fiscal balance).
Capital outlays as percent of GDP, lagged by two years (or CAGDP2) is positively
associated with NGAB, meaning the higher the level of capital spending as percent of
GDP the higher (lower) fiscal balance (budget deficit). This suggests that investments in
capital projects may, in fact, improve fiscal position. A plausible explanation is that
higher capital spending results to higher economic activity which then leads to higher
taxable base and thus higher taxes, and finally lower deficit (or higher fiscal balance).
27

See Diokno [2005] for a a comparative analysis of the two tax reform programs.
See Appendix B.
29
A major exception is debt service because it is automatically appropriated. A higher inflation that leads to
higher nominal interest rates may lead to result to higher aggregate expenditure.
28

12

A. National Government Account Balance (NGAB)

Table 1. Dependent Variable: NGAB as a percent of GDP
Variable
Econgr
Inflat
M3gdprat
Reer
Intgdp
Cagdp2
Iraexp
Taxeffort
Taxref86
Taxref97

Adjusted R2
No. of Observations

0.021
(0.089)
0.050
(0.039)
-0.274**
(0.107)
-0.032
(0.047)
-0.890**
(0.400)
-

0.013
(0.087)
0.047
(0.038)
-0.264**
(0.105)
-0.026
(0.046)
-0.833**
(0.394)
-

0.115
(0.113)
1.274***
(0.259)
IV for taxeffort
IV for taxeffort

0.111
(0.080)
0.074**
(0.032)
-0.252***
(0.079)
-0.031
(0.035)
-0.452
(0.275)
1.008***
(0.333)
0.123
(0.087)
1.282***
(0.192)
IV for taxeffort
IV for taxeffort

0.114
(0.111)
1.231***
(0.256)
IV for taxeffort

0.108
(0.081)
0.072**
(0.032)
-0.248***
(0.080)
-0.029
(0.036)
-0.438
(0.283)
0.996***
(0.338)
0.122
(0.086)
1.268***
(0.206)
IV for taxeffort

0.6562
23

0.7986
23

0.671
23

0.8003
23

*Statistically significant at the 10-percent level, ** statistically significant at the 5-percent level, and
*** statistically significant at the 1-percent level.
IV: Instrumental variable

Tax effort (TAXEFFORT) turned out to be the most robust determinant of fiscal balance.
As expected, it has a positive relationship with NGAB; that is, a higher tax effort is
associated with larger fiscal surplus or lower deficit. The contribution of the 1986 tax
reform program (TAREF86) to tax effort is positive and highly significant while that of
the 1997 tax reform program (TAXREF97) is negative and statistically significant (see
Appendix B).
Economic growth (ECONGR) turned out to be statistically insignificant which suggest
that the national governments fiscal behavior during the period under study is generally
invariant to the economys growth performance. It may be argued that there might be
offsetting effects at work on the revenue and expenditure sides of the fiscal equation. On
the revenue side, the responsiveness of the tax system to GDP growth has increased after
the 1986 tax reform program and declined after the 1997 comprehensive tax reform
program. On the expenditure side, citizens do not have large entitlements such as food

13

subsidy or unemployment insurance, thus there is little mandatory pressure for the
Government to spend more when the economy is slowing down. In fact, external
pressure from external creditors and international financial institutions for the
Government to pursue fiscal consolidation when the economy is slowing down
effectively limits the political leaders countercyclical spending tendency.
Real exchange rate (REER) also turned out to be statistically insignificant. This results
suggest that the effect of a peso depreciation on fiscal stance is unclear, arising plausibly
from offsetting effects of the peso depreciation on revenues and expenditures. The
depreciation (appreciation) of the peso has positive (negative) effect on revenues through
the higher (lower) peso value of imports, other things constant. On the expenditure side,
the depreciation (appreciation) of the peso has negative (positive) effect on debt servicing
since about half of the national government debt are from foreign sources.
B. Consolidated Public Sector Fiscal Position (CPSFP)

Table 2. Dependent Variable: CPSFP as a percent of GDP

Variable
Econgr
Inflat
M3gdprat
Reer
Intgdp
Cagdp2
Iraexp
Taxeffort
Taxref86
Taxref97

Adjusted R2
No. of Observations

1
0.361**
(0.161)
0.161
(0.107)
-0.380**
(0.142)
-0.155
(0.111)
-1.569*
(0.885)
0.231
(0.142)
1.509***
(0.370)
IV for taxeffort
IV for taxeffort

0.6987
18

3
0.288*
(0.155)
0.107
(0.103)
-0.283*
(0.140)
-0.056
(0.113)
-0.813
(0.891)
-

0.324***
(0.098)
0.159**
(0.065)
-0.297***
(0.084)
-0.128*
(0.064)
-0.842
(0.513)
1.468***
(0.350)
0.175*
(0.088)
1.527***
(0.216)
IV for taxeffort
IV for taxeffort

0.200
(0.135)
1.133***
(0.381)
IV for taxeffort

0.281**
(0.099)
0.125*
(0.067)
-0.242**
(0.088)
-0.067
(0.073)
-0.418
(0.562)
1.357***
(0.349)
0.160***
(0.087)
1.288***
(0.255)
IV for taxeffort

0.8849
18

0.7309
18

0.8897
18

*Statistically significant at the 10-percent level, ** statistically significant at the 5-percent level, and
*** statistically significant at the 1-percent level.
IV: Instrumental variable

14

The statistically significant determinants of the consolidated public sector fiscal

position (CPSFP) are the following: economic growth, inflation, domestic liquidity,
capital outlays, intergovernmental fiscal transfers (IRA) and tax effort. Two
variables economic growth and intergovernmental fiscal transfer which were not
statistically significant using the national government fiscal balance as the
explanatory variable for budget deficit turned out to be significant using the
consolidated public sector deficit concept.
Real GDP growth rate (ECONGR) is found to be positively associated with CPSFP,
indicating economic growth may lead to a better fiscal position. The most plausible
explanation for this result is that the other public sector entities the monitored
corporations, the government financial institutions, and social security institutions like
the Government Service Insurance System, the Social Security System, and Philhealth,
and others perform better financially when the economy is growing and poorly when
the economy is slowing down.
Intergovernmental fiscal transfers (IRA) is found to be positively associated with the
consolidated public sector fiscal balance, that is, the higher the IRA, the higher the
consolidated fiscal balance. The explanation is that under existing budget rules, local
governments are mandated by law to generate a surplus of at least 5 percent for
contingency; the higher the grant, the higher the mandated overall mandated surplus for
local governments, and consequently the higher the consolidated fiscal surplus.
C. Summary of Results

First, using NGAB, the narrowest measure of fiscal balance, the statistically significant
determinants are the following: inflation, domestic liquidity, capital outlays, and tax
effort. On the other hand the following variables were found to be statistically
insignificant: economic growth, REER, interest payment as percent of GDP, and
intergovernmental grant (IRA) as percent of total government expenditures.
Second, using CPSFP, the broader measure of fiscal balance, the statistically significant
determinants of fiscal balance are the following: economic growth, inflation, domestic
liquidity, capital outlays, intergovernmental fiscal transfers (IRA) and tax effort. Two
variables economic growth and intergovernmental fiscal transfer which were not
statistically significant using the national government fiscal balance as the explanatory
variable for budget deficit turned out to be significant using the consolidated public
sector deficit concept.
Third, the negative association of domestic liquidity with fiscal balances implies that in
financing the deficit, the government may opt to resort to debt financing first , rather than
printing money or increasing taxes. In the Philippine context, monetizing the deficit is
not a preferred option because of legal restrictions and financial limitations on the

15

monetary authorities (BSP). On the other hand, passing new tax laws to raise revenues
has always been a difficult option politically.
Fourth, tax effort has been the most robust determinant of national government fiscal
balance or the broader measure of fiscal balance (CPSFP). What is more interesting
though is the exploratory regression of tax effort and the tax reform dummy variables. It
is found that tax effort is positively related to the 1986 tax reform at a 1% level of
significance while it is negatively associated with 1997 CTRP at the 15% level of
significance.30 A plausible explanation why the coefficient of the 1997 CTRP is less
significant is that while major reforms initiated in 1986 such as value-added (VAT) are
still in place, the VATs effect was not as potent as before because the tax base was
narrowed as a result of the 1997 CTRP.31
Fifth, real GDP growth rate (ECONGR) is found to be positively associated with fiscal
balance using CPSFP as the explanatory variable, but insignificant if the more limited
deficit concept (NGAB) is used. The results suggests that strong economic growth may
lead to a better fiscal position. While the effect of economic growth on the national
government deficit is unclear, its effect on other public sector entities is unequivocally
positive. The monitored corporations, the government financial institutions, and social
security institutions including Philhealth, local governments perform better financially
when the economy is growing and poorly when the economy is slowing down.
Finally, intergovernmental fiscal transfer (IRA) is found to be positively associated with
the consolidated public sector fiscal balance, though its association with the national
government deficit is found to be statistically insignificant. The empirical result suggests
that the higher the IRA, the higher the consolidated fiscal balance. The explanation is that
under existing budget rules, local governments are mandated by law to generate a surplus
of at least 5 percent to cover future contingencies; the higher the grant, the higher the
mandated overall mandated surplus for local governments, and consequently the higher
the consolidated fiscal surplus.
V. IMPLICATIONS FOR POLICY

Based on the foregoing results and discussion, the following implications for policy
appear warranted. First, in order to arrive at more meaningful decisions, policymakers
should use the broader measure of consolidated public sector fiscal position (CPSFP)
rather than the narrower concept of national government account balance (NGAB) in
evaluating the fiscal health of the government. The empirical results for the regression
using NGAB as the dependent variable suggest that economic growth rate and
intergovernmental fiscal transfers do not affect fiscal balance. But using the broader
30
31

See Appendix B.
Diokno [2005]. The most important are R.A. 8184 An Act Restructuring the Excise Tax on Petroleum
Products; R.A. 8241 Additional exemptions to value-added tax, VAT and more recently R.A. 9010 An
Act Deferring the Imposition of VAT on Certain Services Imposed in R.A. 8761.
16

concept of CPSFP, the results suggest that economic growth rate and intergovernmental
fiscal transfers are both associated with fiscal stance positively.
Second, infrastructure investment has a positive and robust influence on fiscal health.
Consequently, government must prioritize spending on public infrastructure. Investment
in productivity-enhancing capital projects makes private investment more productive,
reduces transactions costs, and increases the profitability of private sector businesses.
Regretfully, because of the shortsightedness of policymakers, spending in public
infrastructure has always suffered cuts during periods of fiscal consolidation.
Finally, tax effort has been the strongest positive determinant of the Philippines fiscal
health. Public policy must be directed to improving revenue effort, not only by
correcting existing weaknesses in the tax system (such, as for example, narrow tax base
because of the proliferation of fiscal incentives laws), but also by improving tax
administration.

17

Appendix A
Data description and source
Variables
NGAB
CPSFP

ECONGR
INFLAT
M3GDPRAT
REER

INTGDP
CAGDP2
IRAEXP
TAXEFFORT
DECENT
TAXREF86
TAXREF97

Description & source

National government account balance: total government revenues less
expenditures, as a percent of GDP. Source: Fiscal Statistics Handbook (FSH),
Department of Budget & Management (DBM).
Consolidated public sector fiscal position: NGAB plus other public sector (i.e.
Central Bank (CB) restructuring accounts, major non-financial government
corporations (GOCC), government financial institutions (GFI), local government
units (LGU), social security institutions, the Oil Price Stabilization fund & the
Bangko Sentral ng Pilipinas (BSP)), as a percent of GDP. Source: FSH, DBM
Growth rate of real GDP. Source: FSH, DBM
Inflation rate: rate of growth of the consumer price index (2000=100). Source:
FSH, DBM.
The ratio of liquid liabilities of the financial system (M3) to GDP. Source: Bangko
Sentral ng Pilipinas (BSP) & FSH, DBM..
REER IS the nominal effective exchange rate of the peso adjusted for inflation
rate differentials with the countries whose currencies comprise the nominal
effective exchange rate basket (NEER1). Formally, REER = NEER1 x
(domestic price index / weighted price index of countries whose currencies are
in the NEER1 basket). NEER1 Is the weighted average exchange rate of the
peso vis--vis a basket of foreign currencies (i.e. US Dollar, Japanese Yen,
European Monetary Unit (Euro), British Pound) unadjusted for the effects of
inflation. Source: (BSP).
National government debt service interest payments as a percent of GDP.
Source: FSH, DBM.
Two-year lag of Capital Outlay expenditures as a percent of GDP. Source:
FSH, DBM.
Total internal revenue allotment (IRA) for year t as a percent of GDP. Source:
FSH, DBM.
Total tax revenue as a percent of GDP. Source: FSH, DBM.
Dummy variable indicating years in which the 1991 Local Government Code of
the Philippines (LGC) is in effect, implemented in 1992.
Dummy variable indicating years in which the 1986 Tax Reform Program (TRP)
is in effect, implemented in 1987.
Dummy variable indicating years in which R.A. 8424, the 1997 Comprehensive
Tax Reform Program (CTRP), was in effect.

18

Appendix B
DEPENDENT VARIABLE: Taxeffort

Variable
Taxref86
Taxref97
Econgr
Inflat

4.468****
(0.740)
-1.101*
(0.740)
-

4.657****
(0.814)
-1.128*
(0.754)
-0.046
(0.076)
-

4.639****
(0.817)
-1.322*
(0.784)
-0.099
(0.095)
-0.041
(0.044)

0.6007
23

0.5980
23

Adjusted R2
0.6132
No of Observations
23
Level of Significance: *15%, ** 10%, *** 5%, **** 1%

19

References
Aizenman, Joshua and Ilan Noy (2003), Endogenous Financial Openness: Efficiency
and Political Economy Considerations, National Bureau of Economic Research
(NBER), Working Paper No. 10144.
Auerbach, Alan (2003), Fiscal Policy, Past and Present, National Bureau of
Economic Research, Working Paper No. 10022.
Bayraktar, Nihal and Yan Wang (2006), Banking Sector Openness and Economic
Growth, World Bank Working Series Paper No. 4019.
Clarida, Richard and Joe Prendergrast (1999), Fiscal Stance and the Real Exchange
Rate: Some Empirical Estimates, NBER Working Paper No. 7077.
de Dios, Emmanuel S., Benjamin E. Diokno, Emmanuel F. Esguerra, Raul V. Fabella,
Ma. Socorro Gochoco-Bautista, Felipe M. Medalla, Solita C. Monsod, Ernesto M.
Pernia, Renato E. Reside, Jr., Gerardo P. Sicat, and Edita A. Tan (2004), The
Deepening Crisis: The Real Score on Deficits and The Public Debt, Discussion
Paper No. 0409, University of the Philippines School of Economics.
Denizer, Cevdet, Raj M. Desai and Nikolay Gueorguiev (1988). The Political Economy
of Financial Repression in Transition Economies, NBER, Working Paper No.
2030.
Department of Budget and Management. Fiscal Statistics Handbook, 1980-1994.
__________. Fiscal Statistics Handbook, 1984-2003.
Diokno, Benjamin (2005). Reforming the Tax System: Lesson from Two Tax Reform
Programs Discussion Paper No. 0502. University of the Philippines School of
Economics, unpublished.
__________ (1999). Beyond the Basics: The Philippine Case, in Schiavo-Campo, S
(Ed.) Governance, Corruption and Public Financial Management. Philippines:
Asian Development Bank, pp. 135-42.
__________ (1995). Fiscal Policy for Sustainable Growth, in R. V. Fabella and H.
Sakai (eds.) Towards Sustained Economic Growth. Tokyo: Institute of
Developing Economies, pp. 33-56.
Dornbusch, Rudiger, Stanley Fischer and Richard Startz (2003). Macroeconomics, 9th
Ed. USA: McGraw-Hill, Inc.
Duo Qin, Marie Anne Cagas, Geoffrey Ducanes, Nedelyn Magtibay-Ramos and Pilipinas
Quising (2005). Empirical Assessment of Sustainability and Feasibility of

20

Government Debt: The Philippine Case, Economic and Research Department

Working Paper Series No. 64. Philippines: Asian Development Bank.
Easterly, William and Klaus Schmidt-Hebbel (1994). Fiscal Adjustment and
Macroeconomic Performance: A Synthesis, in Easterly, W.. C. A. Rodriguez and
K. Schmidt-Hebbel (eds.) Public Sector Deficits and Macroeconomic
Performance. New York: Oxford University Press, pp. 15-78.
Edwards, Sebastian (1996). Public Sector Deficits and Macroeconomic Stability in
Developing Economies, NBER, Working Paper No. 5407.
Manasan, Rosario (1998). Fiscal Reform and Economic Development, in Economic
Development Management Asia and the Pacific (EDAP) Joint Policy Studies No.
2, Public Finance Reform in the Asia Pacific. Korea: The Korea Development
Institute, pp. 223-262.
Mankiw, N. Gregory (2000). Macroeconomics, 4th Ed. New York: Worth
Publishers.
National Statistical Coordination Board (NSCB). Philippine Statistics Yearbook (various
years).
Obstfeld, Maurice and Kenneth Rogoff (1996). Foundations of International
Macroeconomics. Massachussets: The Massachussets Institute of Technology
Press.
Paderanga, Cayetano W. Jr. (2004). The Philippines, in A. Kohsaka (ed.) Fiscal Deficits
in the Pacific Region. London: Routledge, pp. 301-321.
Peach, Richarad and Charles Steindel (2000). Summary of Observations and
Recommendations on Alan Auerbach presentations at the Federal Reserve Bank
of New York. Economic Policy Review.
Persson, Torsten and Guido Tabellini (1998). Political Economics and Public Finance,
[prepared for the handbook of Public Economics, Vol III, ed. By Alan Auerbach
and Martin Feldstein, preliminary draft]
Romer, David (2001). Advanced Macroeconomics, 2nd Ed. New York: McGraw-Hill
Higher Education.
Sachs, Jeffrey and Noriel Roubini (1988). Political and Economic Determinants of
Budget Deficits in the Industrial Democracies. NBER Working Paper No. 2862.
Saez, Emmanuel (1999). The Effect of Marginal Tax Rates on Income: A Panel Study
of Bracket Creep, NBER, Working Paper No. 7367.

21

Sicat, Gerardo (2003). Economics, New Edition. Volumes 1-3. Philippines: Anvil
Publishing, Inc.
World Bank (2005). Philippines: Meeting Infrastructure Challenges. Philippines: The
World Bank.
Woo, Jaejoon (2001). Economic, Political and Institutional Determinants of Public
Deficits, Journal of Public Economics, 87, 387-426.

22